HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Heidler, Konrad, Mou, Lichao, Baumhoer, Celia, Dietz, Andreas, Zhu, Xiao Xiang
Format: Preprint
Veröffentlicht: 2021
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866916454686785536
author Heidler, Konrad
Mou, Lichao
Baumhoer, Celia
Dietz, Andreas
Zhu, Xiao Xiang
author_facet Heidler, Konrad
Mou, Lichao
Baumhoer, Celia
Dietz, Andreas
Zhu, Xiao Xiang
contents Deep learning-based coastline detection algorithms have begun to outshine traditional statistical methods in recent years. However, they are usually trained only as single-purpose models to either segment land and water or delineate the coastline. In contrast to this, a human annotator will usually keep a mental map of both segmentation and delineation when performing manual coastline detection. To take into account this task duality, we therefore devise a new model to unite these two approaches in a deep learning model. By taking inspiration from the main building blocks of a semantic segmentation framework (UNet) and an edge detection framework (HED), both tasks are combined in a natural way. Training is made efficient by employing deep supervision on side predictions at multiple resolutions. Finally, a hierarchical attention mechanism is introduced to adaptively merge these multiscale predictions into the final model output. The advantages of this approach over other traditional and deep learning-based methods for coastline detection are demonstrated on a dataset of Sentinel-1 imagery covering parts of the Antarctic coast, where coastline detection is notoriously difficult. An implementation of our method is available at \url{https://github.com/khdlr/HED-UNet}.
format Preprint
id arxiv_https___arxiv_org_abs_2103_01849
institution arXiv
publishDate 2021
record_format arxiv
spellingShingle HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline
Heidler, Konrad
Mou, Lichao
Baumhoer, Celia
Dietz, Andreas
Zhu, Xiao Xiang
Computer Vision and Pattern Recognition
Image and Video Processing
Deep learning-based coastline detection algorithms have begun to outshine traditional statistical methods in recent years. However, they are usually trained only as single-purpose models to either segment land and water or delineate the coastline. In contrast to this, a human annotator will usually keep a mental map of both segmentation and delineation when performing manual coastline detection. To take into account this task duality, we therefore devise a new model to unite these two approaches in a deep learning model. By taking inspiration from the main building blocks of a semantic segmentation framework (UNet) and an edge detection framework (HED), both tasks are combined in a natural way. Training is made efficient by employing deep supervision on side predictions at multiple resolutions. Finally, a hierarchical attention mechanism is introduced to adaptively merge these multiscale predictions into the final model output. The advantages of this approach over other traditional and deep learning-based methods for coastline detection are demonstrated on a dataset of Sentinel-1 imagery covering parts of the Antarctic coast, where coastline detection is notoriously difficult. An implementation of our method is available at \url{https://github.com/khdlr/HED-UNet}.
title HED-UNet: Combined Segmentation and Edge Detection for Monitoring the Antarctic Coastline
topic Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2103.01849